Support Vector Machine Implementation for Classifying Public Sentiment on Electronic Stamps Use in Civil Servant Registration

Aug 29, 2026

Upload by Paura

Civil servants are central to government administration, while the recruitment of civil servant candidates is a critical stage with long-term effects on organizational effectiveness. In Indonesia, digital electronic stamps (e-stamps) have been introduced for CPNS registration to improve efficiency and transparency, yet public reactions on Twitter/X are mixed, including positive views on digitalization and reports of technical difficulties. Sentiment analysis is therefore required. Using a Twitter/X dataset of 1,249 reviews, this study applies Support Vector Machine (SVM) with hyperparameter tuning via GridSearchCV to classify sentiments, achieving an accuracy of 92% with cost=100 and gamma=0.01.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Introduction

Theoretical Framework

Sentiment Analysis Concepts

Text Mining and Preprocessing

Feature Selection for Text Classification

Volume 13 Issue 05 May 2025, Page no.– 5253-5259 Index Copernicus ICV: 57.55, Impact Factor: 8.615 DOI: 10.47191/ijmcr/v13i5.17 Support Vector Machine Implementation for Classifying Public Sentiment on Electronic Stamps use in Civil Servant Registration Nilna Adiba Kamal1, Suparti2, Puspita Kartikasari3 1,2,3 Department of Statistics, Diponegoro University, Indonesia | ARTICLE INFO | ABSTRACT | | --- | --- | | Published Online:
29 May 2025 | Civil Servants play a crucial role in government administration and national development. The selection process for Civil Servant Candidates is a critical stage in civil servant management, as it has long-term implications for organizational effectiveness. With the advancement of digital technology, the Indonesian government has implemented electronic stamps in the civil servant candidate registration process to enhance efficiency and transparency. However, this policy has received various responses from the public, particularly on the social media platform Twitter. Many users express positive opinions regarding the convenience of administrative digitalization, while others report technical issues in using electronic stamps. Therefore, sentiment analysis is needed to understand public responses to this policy. One of the effective methods for sentiment classification is Support Vector Machine (SVM), which can optimally separate positive and negative opinions. This study utilizes a dataset comprising 1,249 reviews collected from Twitter/X. The best SVM model is selected through hyperparameter tuning using the GridSearchCV technique. The findings indicate that the SVM model with cost = 100 and | | Corresponding Author:
Suparti | gamma = 0.01 achieves the best performance, with an accuracy of 92%, precision of 86.48%, recall of 86.48%, F1-score of 86.48%, and a Kappa-Statistic of 81% . | | KEYWORDS: Civil Servants; E-meterai; Sentiment Analysis; Twitter/X; Support Vector Machine | | I. INTRODUCTION The role of Civil Servants (PNS) is crucial in supporting the country's administration and development. The recruitment process of Civil Servant Candidates (CPNS) is an important stage because it has a long-term impact on the effectiveness of public organizations. Along with the advancement of digital technology, the Indonesian government implemented estamps in the CPNS registration process to increase administrative efficiency and transparency. Although the electronic stamps policy aims to increase the transparency and efficiency of CPNS registration, its implementation has triggered various responses on social media, especially Twitter/X. As a platform that represents real-time public discussions, Twitter/X is often used in sentiment analysis studies because it can record opinions along with time and location [1] . According to Katadata, Indonesia is the fourth largest Twitter/X user globally, with 27.5 million users. The public has expressed various opinions via Twitter/X since the implementation of electronic stamps, ranging from support for digitalization to technical complaints, such as failure to access and affix electronic stamps. These problems are considered to hinder the registration process and cause losses. Therefore, sentiment analysis is needed to understand the public response, by categorizing opinions into positive and negative sentiments. Sentiment analysis is a technique for classifying public opinion into positive or negative sentiment [2] . Research by Herwinsyah and Witanti [3], shows that SVM can classify sentiment related to COVID-19 vaccination with 89% accuracy. Ratino et al. [4], also found that SVM is superior to NaïveIBayes, with accuracies of 80.23% and 78.02%, respectively. Meanwhile, research by Yusupa and Tarigan [5] showed that SVM is quite reliable in analyzing public sentiment towards electric vehicles in Indonesia, with an accuracy of 75.62% . This research was conducted to develop an effective classification model utilizing the

FAQs

Why is sentiment analysis needed for e-stamps policy in CPNS registration?

The e-stamps policy aims to improve efficiency and transparency, but its rollout triggers varied responses on Twitter/X, including positive opinions and technical complaints. Sentiment analysis helps categorize public opinions into positive and negative sentiments objectively.

What dataset and modeling approach are used in this study?

The study uses 1,249 Twitter/X reviews and applies Support Vector Machine (SVM) for sentiment classification. The best model is selected using hyperparameter tuning with GridSearchCV.

What performance results does the best SVM model achieve?

The selected SVM configuration is cost=100 and gamma=0.01. It achieves 92% accuracy, with precision, recall, and F1-score reported at 86.48%, and a Kappa-Statistic of 81%.

Aug 29, 2026

Upload by Paura